Let's Learn from a Loser: How I Lost $10k Outsourcing

[mememan]

The short is that I'm out about $10k through outsourcing. I hope that sharing my mistakes will help others avoid similar roadblocks and allow me to reflect as I try to get my ideas made once again.

I've honestly struggled with typing this message a few times since it always ended up being too long for a first post. No reason to hit you all with a wall of text if it doesn't help you. It's a better idea to craft responses based on interest anyway.

I've run the gamut in terms of developers: domestic and offshore, individuals and firms. My tally of failed attempts stands at 3 at the moment, and each time I've lost consecutively more money.

I am trying to get a very simple site built. It uses a single aggregator API and is essentially a price comparison affiliate site. There are a ton of competitors out there, but it complements other ideas I have and it's basic enough that it should be a good starter job for both me and a developer.

However, that hasn't been the case. Incoming is a lot of "I" statements, but I'm not doing it as a "woe is me" sort of thing. I know I've screwed up and just want to show that you can and more than likely will screw up even if you're doing your homework. To wit:

- I've researched how to outsource successfully in detail. Articles, podcasts, videos, and more.
- I've crafted job listings in a variety of ways and attempted searching and inviting people.
- I've prepared specs, mockups, and even videos explaining my job.
- I've found codebases we can utilize.
- I've offered hourly jobs at high rates and decent budgets.
- I've interviewed via Skype, email, and phone.

In other words, I feel that I've done what I can to make this project a success with the exception of coding it myself. Of course, I know that's not 100% true and I have made mistakes along the way that have contributed to the site's current state, or lack thereof.

The last two developers I worked with were $40 and $50 an hour and from Eastern Europe. The 1st had more than 5k completed hours on oDesk. The 2nd had excellent communication at the start, but it became evident as time went on that he was a "black hole" developer who would suck time and money forever. The one before these two was a domestic firm/team that was just a nightmare overall.

Anyway, here are some of my main takeaways at the moment:

- When you put a budget down, that is where negotiation will start. $1k-$5k is essentially putting down $5k+.
- No payments outside of an escrow system like eLance. If you offer to pay the platform fees in full and and they won't work on the site, say adios to them.
- NEVER use anything but an American Express credit card if there is even the slightest chance you may want your money back. Tie this with a PayPal transaction for a 2nd level of protection if you can. MC and Visa are absolute crap for protection.
- Don't make a decision when you only have one provider you like. You need at least 2-3 you are absolutely in LOVE with. It's a cliche, but go with your gut.
- Development discounts are suspicious.
- Don't sign anything, especially with a US based firm.
- For the love of god, NEVER go hourly. It's a lot harder to chargeback 10 hours or so a week for 10 weeks than it is a lump sum payment. Additionally, assume the people at your credit card company are idiots. After almost a year I had a chargeback reversed with no explanation.
- Don't sit around on any job for more than 2 weeks. 80 hours or less should be enough time to have something basic and actionable.
- Fire quickly. Leave NO feedback on small jobs if you're going to work with them on something else. Your only bargaining chips are money and feedback.
- Most agencies/firms suck domestic or abroad. There are too many places for your needs to get screwed up between sales and the developer.
- Hourly rate means nothing.

Dane Maxwell recently had an AMA on here. On post #29, he had a list of 15 things that are deal breakers with developers. It really resonated with me since it defined what I had been looking for in a developer so clearly.

My overall experience is that, on average, developers like to badmouth a lot of sensible things: code commenting, paper prototyping, in other words, anything that results in them coding less.

I look online and I see all these places that have cheap development resources and assets. AppSumo, ThemeForest, WrapBootstrap, etc. I've come to the conclusion that my dream developer will be someone who can take these resources and a week or two and build a basic, usable version of this site.

Until I find someone who can do that, my wallet is closed, and I would advise anyone looking to do a MVP to seriously consider doing the same until you can find a single developer or at most a very small team that already has licenses to several of the above mentioned resources and understands that custom doesn't necessarily mean better.

Any critiques, feedback, comments, or questions, feel free to post.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[brycer9]

Thanks for sharing! Btw, there is a great conference call on the INSIDERS about outsourcing on Elance. It probably would have saved you 10k >.>

 

[masaldana2]

Hey men you're not a loser!

 

[dknise]

https://www.thefastlaneforum.com/sh...ram-is-STUPID!?highlight=learning+code+stupid

The long standing debate! To outsource or not to outsource?

As a coder myself, I've been very vocal about the problems you faced over in that thread.

- I've researched how to outsource successfully in detail. Articles, podcasts, videos, and more.
- I've crafted job listings in a variety of ways and attempted searching and inviting people.
- I've prepared specs, mockups, and even videos explaining my job.
- I've found codebases we can utilize.
- I've offered hourly jobs at high rates and decent budgets.
- I've interviewed via Skype, email, and phone.

And yet regardless of all of this...

- Outsourcers do not care if you succeed.
- Outsourcers do not care about the quality of your product.
- Outsourcers, if paid based on time, will take as long as possible.
- Outsourcers do not care about code quality. If it functions they get paid.

Of course, there are exceptions to this. I've contracted on so many projects for extra cash, but I also think like an entrepreneur and feel a responsibility to provide a quality product for the price they pay.

I hope you get your product built out and out there! You've already taken 10x the amount of steps most people contracting out a job will take.

Goodluck!

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[mememan]

Thanks for sharing! Btw, there is a great conference call on the INSIDERS about outsourcing on Elance. It probably would have saved you 10k >.>

I actually signed up for INSIDERS's just for that call. I listened to it and while I got confirmation on some of the methods I'd been using, I didn't really hear anything that I didn't already know. If someone is completely new to outsourcing, though, it is definitely worth a listen.

 

[mememan]

I'll have to go through and read that post in its entirety then

You're absolutely right though, and it doesn't matter if your budget is $2k or $20k. I've talked with at least 100-200 developers and I've never once had someone willing to cut the crap and say, "I can charge you $1k, or you can buy this for $50 and we'll get to work."

The worst thing is that the whole time there was been an opensource codebase on GitHub for this exact project. No one will do anything with it, I suspect since that's just taking away work from them.

Thanks for the support. I'm willing to answer any particular questions too!

 

[dknise]

I'll have to go through and read that post in its entirety then

It's a jumbled mess of back and forths!

You're absolutely right though, and it doesn't matter if your budget is $2k or $20k. I've talked with at least 100-200 developers and I've never once had someone willing to cut the crap and say, "I can charge you $1k, or you can buy this for $50 and we'll get to work."

Very true. Usually when I take a contract, I'll do so based on what it requires and what I already have built. A few months ago I walked in to a company on a Sunday to automate some accounting stuff for $5k and walked out with it fully completed with way more features than they asked for in 7 hours. I was confident I could get it done in less than a day and we came to an agreed price, but after they saw how little I actually had to do they were furious and wanted to only pay about $200. I ended up coming down to $2500, but they're STILL mad about it. If you ask me "can you do this for $5k" and I know I can do it in several hours, I'm going to agree to do it for $5k hahaha.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[tafy]

A few months ago I walked in to a company on a Sunday to automate some accounting stuff for $5k and walked out with it fully completed with way more features than they asked for in 7 hours

That was a rookie mistake, and ofcourse they were mad as they got ripped off. Contractors often do jobs based on price and if it seems it can be done way faster than the client thought they always stretch it out till atleast 3/4 of the time to deliver a bit early. Thats why I always hire builders and electricians etc on hourly or daily rate. Also buy in the materials they need so they cant overcharge.

 

[exige]

That was a rookie mistake, and ofcourse they were mad as they got ripped off. Contractors often do jobs based on price and if it seems it can be done way faster than the client thought they always stretch it out till atleast 3/4 of the time to deliver a bit early. Thats why I always hire builders and electricians etc on hourly or daily rate. Also buy in the materials they need so they cant overcharge.

They didn't get ripped off, and its not a rookie mistake. It's value based consulting pricing. Its just that to do value based pricing successfully is harder, but with the right selling and after-sale approach its not impossible. In this case his client got their accounting stuff automated. What's the ROI and time-to-realize that ROI on that for them? It could be saving them way more than $5k.

 

[tafy]

They might not have been ripped off but they certainly felt ripped off! You had to lower your price to $2,500?

If you thought $5k was fair then why not fight for the full amount?

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[exige]

To the OP, it sounds like you might be more successful if you hired a solution designer/architect to spend a few hours and put together a design and development backlog based around the available off-the-shelf components, and then let the developer(s) work off of that. Not using the available code you found could be NIH syndrome, or there could be something wrong with it that the devs see.

I do agree that 2 weeks is enough time to see something actionable though. After a sprint you should see _something_ working.

There's a big difference between a code monkey and a solution developer that really knows his stuff. It could be that you are handing something to code monkeys that really needs a little TLC from somebody more experienced and with a different skill set. It may also be that you have unrealistic expectations for how much it will really cost to build what you want.

 

[mememan]

but after they saw how little I actually had to do they were furious and wanted to only pay about $200. I ended up coming down to $2500, but they're STILL mad about it

I'm definitely in the wrong field lol. I'm not sure if the $200 is actually $2000, but the agreed upon $350 an hour rate sounds great.

I'd say this is a great example of exactly why you always low-ball your budget. On eLance, the prefixed bid ranges are less than $500, $500-$1k, $1k-$5k and others. On my project I've selected the last one a few times since, for the right person, I'd be willing to go up to $3k-$5k for implementing all the features I have planned. However, what you end up getting is people who hover around $5k and up and adjust their proposal timeline for this fact.

Also, know what resources are out there already on the cheap. With the exception of a logo, I'd say there is very little on the graphics/UI end that needs to be customized. You can mash together elements of ThemeForest products (especially if you get the raw PSD files). Even better, there are a lot of free "framework" design tools that have buttons, icons, and other UI elements that can be assembled with a little more freedom. Even on the programming end, there is no reason to just code from scratch all the time, but no programmer will be willing to look at what's out there to save you hundreds or thousands.

 

[mememan]

Contractors often do jobs based on price and if it seems it can be done way faster than the client thought they always stretch it out till atleast 3/4 of the time to deliver a bit early. Thats why I always hire builders and electricians etc on hourly or daily rate. Also buy in the materials they need so they cant overcharge.

I think this is something you have to play by ear. At this point, I know that I'll really try to avoid hourly projects in the future, at least when it comes to digital work.

On my project, for example, the last developer I had started with a $2k fixed rate. I was already out $3k from the developer before him and this was a little higher than what I wanted, but still acceptable. He then wanted to switch to hourly since the last project he did had overages. Almost everything online written from the programmer's perspective says the decent ones will only work hourly, so I went ahead and did this.

The contract ballooned to around $5k, mainly since I just didn't say "enough" and eat my losses earlier. It didn't help that this developer had far better communication than almost anyone else I'd spoken with up to this point, domestic or otherwise.

With hourly, at least my limited experience, you'll see 2 things you really won't like, mainly (a) them learning and looking things up on your dime and (b) being charged for communications (email, text chat, etc). Both of these can be inflated without your knowing, though of course that can happen with the actual coding too.

To the OP, it sounds like you might be more successful if you hired a solution designer/architect to spend a few hours and put together a design and development backlog based around the available off-the-shelf components, and then let the developer(s) work off of that. Not using the available code you found could be NIH syndrome, or there could be something wrong with it that the devs see.

I do agree that 2 weeks is enough time to see something actionable though. After a sprint you should see _something_ working.

There's a big difference between a code monkey and a solution developer that really knows his stuff. It could be that you are handing something to code monkeys that really needs a little TLC from somebody more experienced and with a different skill set. It may also be that you have unrealistic expectations for how much it will really cost to build what you want.

Absolutely agree. However, I've had everything structured for these last 2 developers. That doesn't mean they've utilized it though.

Looking at that INSIDERS's "How to Hire on eLance" proposal, I've prepared something similar to that and a lot more. I'm honestly burnt out presenting the same materials in different fashions especially since this was never my core business plan.

I know my prior cost estimates were definitely doable. I could actually have just bought a preexisting site in the exact same space for less than I've spent so far.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Nick]

I've talked with at least 100-200 developers and I've never once had someone willing to cut the crap and say, "I can charge you $1k, or you can buy this for $50 and we'll get to work."

The worst thing is that the whole time there was been an opensource codebase on GitHub for this exact project. No one will do anything with it, I suspect since that's just taking away work from them.

Instead of talking to developers and expecting them to research open source projects that will save you thousands of dollars on your project for free (I hope you are not under the misguided assumption that developers knows by heart all the open source source codes available) try paying them to perform a market research / feasibility check on your project and present you the best options you have (open source code bases have their disadvantages too, you need to be aware of them before deciding to build on them).

 

[exige]

TTTDodder, why don't you post a thread on INSIDERS with a little more detail if you're comfortable. Some of the experienced software guys might be able to spot/point out what it is that seems to be mistifyingly challenging about your project and help de-mistify it.

 

[The-J]

For the love of god, NEVER go hourly. It's a lot harder to chargeback 10 hours or so a week for 10 weeks than it is a lump sum payment. Additionally, assume the people at your credit card company are idiots. After almost a year I had a chargeback reversed with no explanation.

It's funny, because I've known a lot of devs who stopped doing hourly rates and went with retainer rates instead. They hate hourly as much as you do.

dknise has some good advice (because he speaks from the other side). Developers (or any other outsourced technical worker) don't give a F*ck and will gladly take on projects that they think will fail. They don't care because they get paid anyway. They'll never take equity because they know it's a gamble. They just want to pay their bills and eat.

Honestly, I'd go with retainer rates. Better to pay $3k/month than $30/hr, because you never know what bullshit they're gonna charge for.

Also, I gave you speed+ for F*cking up and learning a lot, as well as sharing it with the grateful members of the Fastlane Forum.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[mememan]

TTTDodder, why don't you post a thread on INSIDERS with a little more detail if you're comfortable. Some of the experienced software guys might be able to spot/point out what it is that seems to be mistifyingly challenging about your project and help de-mistify it.

I'll do that. It honestly isn't a very unique or killer idea. I'm not expecting it to make a mint, but I think I could build it up to at least $30k a year. I have revenue and traffic figures for two sites that were on Flippa in the niche and I was in touch with one person who was selling a very similar site in the last year.

try paying them to perform a market research / feasibility check on your project and present you the best options you have

I've done that before. I think the problem for me is that people on these outsourcing sites just don't look at things like AppSumo and the Envato marketplaces. I also have the open source tools in mind, I'm not expecting them to know everything

Developers (or any other outsourced technical worker) don't give a F*ck and will gladly take on projects that they think will fail. They don't care because they get paid anyway. They'll never take equity because they know it's a gamble. They just want to pay their bills and eat.

Honestly, I'd go with retainer rates. Better to pay $3k/month than $30/hr, because you never know what bullshit they're gonna charge for.

If anyone reads this, this is the main thing to get out of it. Hourly is practically asking for problems. eLance and other sites like it since it's cheaper and lower risk for them.

With hourly, they have your credit card information and charge you for hours through the WorkView application weekly automatically. There are a few days as a "review period," but honestly, if you don't have the code right in front of you, you can't tell if the work is satisfactory. In any event, the freelancer is guaranteed payment as long as they use the WorkView application. If you dispute and the reviewer see something on the screenshots in question, you'll lose. All disputes are handled by outsourced workers anyway at lowish rates (I've seen oDesk paying sub $5 for these people), so it's not like you have a programming expert checking the job.

Escrow really isn't that great either. First of all, eLance charges $400 for a dispute, so it's not like it is cost effective for most small projects anyway. The dispute arbitration process is supposed to be a joke from what I've read online. But if you want to throw away more money, it's there, and you could probably strongarm the freelancer to at least provide a partial refund with a threat of awful feedback.

There's always the credit card chargeback option if you're willing to nuke your account. It's against the site terms of use, but if you've lost enough, and you don't care, then maybe try that. I'd only do that if you had American Express though.

Here's why. For the first developer I had, they didn't do anything. Literally, they turned in my files and logos to the merchant processor as proof that they had completed the work. I had letters for an attorney confirming they were just my materials, as well as a letter from someone who was a contractor with the firm confirming the same. One year later, after I had been issued the credit and been paid a check previously, the chargeback was "disputed" saying the work had been completed since they had just now got around to actually looking at the case.

 

[healthstatus]

The worst thing is that the whole time there was been an opensource codebase on GitHub for this exact project. No one will do anything with it, I suspect since that's just taking away work from them.

So why haven't you contacted the opensource developers and offered to pay them for your customized version?

 

[dknise]

That was a rookie mistake, and ofcourse they were mad as they got ripped off. Contractors often do jobs based on price and if it seems it can be done way faster than the client thought they always stretch it out till atleast 3/4 of the time to deliver a bit early. Thats why I always hire builders and electricians etc on hourly or daily rate. Also buy in the materials they need so they cant overcharge.

...

They might not have been ripped off but they certainly felt ripped off! You had to lower your price to $2,500?

If you thought $5k was fair then why not fight for the full amount?

As exige said, if you paid me $5k to automate a money saver, would you prefer I stretch it out 3 weeks and lie to you while you keep losing money? I have an ASAP policy that I stick to.

I didn't fight for the full amount because I was suggested to them through a mutual friend. Out of respect for the friend I was willing to negotiate.

They didn't get ripped off, and its not a rookie mistake. It's value based consulting pricing. Its just that to do value based pricing successfully is harder, but with the right selling and after-sale approach its not impossible. In this case his client got their accounting stuff automated. What's the ROI and time-to-realize that ROI on that for them? It could be saving them way more than $5k.

BINGO. In that specific case I fully eliminated two full time jobs and lightened the payload on a 3rd employee. Yearly savings from the project will be around $90k. If any other development team had done the project, they would have pushed for half the yearly savings and taken a month to complete it (and seeing others automation attempts I am 100% positive I'm providing more features and better quality).



This does bring up another huge outsourcing issue everyone should be aware of! I, like many people, am not writing software FOR you. I am LICENSING you the right to use the software I write. This distinction allows me to abstract automation principles out to pre-built packages that I can reuse over and over again on projects. When I coded custom CMS software using WinForms for my dad's business, I did not write custom CMS software using WinForms, I wrote software that generates custom CMS software from a database. Now when I see a need for custom WinForms CMS software from a database, I know I can generate the often 100,000+ lines of base code automatically in the blink of an eye once I know the data structure. Other popular requests are reading from RSS feeds for sales info, parsing email's for info, transferring data in and out of MS products like Excel, and scheduling tasks using windows services. If I ever took an automation contract, you bet I have it included in my C#, Java, and PHP com.averscer packages!

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[mememan]

So why haven't you contacted the opensource developers and offered to pay them for your customized version?

I did. They put the code up for free since they abandoned the project and moved onto full-time jobs.

 

[Worldisyours]

Ok, My solution to you is this.

If you really really really think your site will shoot off.

Move to a free trade country with decent infrastructure in the Asia. Open a company for nothing, like a room as main office, hire a FULL TIME web developer by salary( 1000usd a month is prob 5X the average salary for the countries). He will be loyal. Forever. Make your site. The best one in a year for 12000$. His job is the site now.

Offcourse, you have to move. Now the question is do you feel this site is IT!. Are you willing to sacrifice to get it done. Well, if not then you shouldn't bother anyways.

I think the issue is not you are getting ripped by developers. Its that you are not looking in th eright ways.

All the best of luck.

 

[mememan]

Ok, My solution to you is this.

If you really really really think your site will shoot off.

Move to a free trade country with decent infrastructure in the Asia. Open a company for nothing, like a room as main office, hire a FULL TIME web developer by salary( 1000usd a month is prob 5X the average salary for the countries). He will be loyal. Forever. Make your site. The best one in a year for 12000$. His job is the site now.

Offcourse, you have to move. Now the question is do you feel this site is IT!. Are you willing to sacrifice to get it done. Well, if not then you shouldn't bother anyways.

I think the issue is not you are getting ripped by developers. Its that you are not looking in th eright ways.

All the best of luck.

Haha, thanks for the tip. Honestly, the site idea isn't that good, but hey, if those are the rates I should be paying, then I *KNOW* that I can be a cheapass this next go around on this.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[FastNAwesome]

Just want to share a few views as a developer, hopefully it helps you guys.

1. Outsourcing is like picking a car mechanic. Always a gamble! What you can do is:

- Stack the odds in your favor by performing all due diligence already mentioned in many threads
- Check if they've done related work previously (or will be learning on your dime)
- If they did show you previous related work, ask for proof of credits, recommendations, talk to previous employers
- Ideally you want to find someone who is a good professional and a good person too. Because, many freelancers, just like many car mechanics, when they spot someone not versed enough, will always find reasons to milk more money, and will try to tell you all kinds of stories.
The more you know about web technology, the less they can full you. And start learning about cars too, I know I did and it saves me money


2. Don't Post Your Budget:

- While I browse freelance sites looking for work, I see this all the time. If you're willing to pay $3-5k for something, then - just like OP experienced, you'll get a ton of bids for $5k and for even more. Open another job, same requirements, but with budget $750-1500, and guess what happens? Open third same job, budget $30-250, and guess what happens?

The truth is, too many bidders don't even read or care about your project, they just want to know how much you will pay.

Another extreme is if your budget is too low for something. Good workers will skip it.

Until you gain experience for setting the right budget, let the freelancers give their estimates.

3. Avoid hourly billing:

- It puts the employer under stress and uncertainty. "How much will this actually project cost".

- It's really tricky for the freelancer. So the faster he can do the work - the less he will be paid?

- To make fixed pricing work, you need to be as specific as possible what is the scope of work, so that both you and the worker know when it's finished. And also need to be fair in terms of adding new stuff. It's ok to ask for some small features. But if you keep coming up with new ideas and requirements - be fair and offer an extra pay for that.

4. Keep the pressure: You want regular updates. Set the tempo. Ask about the progress, ask what's done, how things are going, ask to see results. Depending on the project, daily or weekly updates may be more appropriate. But remember that you are not merely a customer here, you are a project manager, so manage workers.

 

[jazb]

This is the only thing putting me down about fastlane businesses - the online game.

I think we are all better off learning to code ourselves. where will we be in 15 years time? perhaps starting another business? oh no, i have to pay someone to make an app for me again, i should of learned to code at the start.

'best time to plant a tree was 20 years ago, second best time is now'

 

[SteveO]

where will we be in 15 years time?

By that time you will be learning to code again as the technology will have changed.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[snowbank]

By that time you will not have to code again as the technology will have changed.

FYP

 

[IAmTheJeff]

By that time you will be learning to code again as the technology will have changed.

I bet that applies to 3 years from now.

 

[SteveO]

FYP

You think that coding will go away? Certainly there have been a lot of advances but do you think putting websites up has become less complicated in the last 10 years?

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[IAmTheJeff]

I can definitively say that throwing something on the Internet is easier now than 10 years ago. What's gotten more difficult is getting your site seen. Ten years ago if you had a website selling purses, you might have been able to get purses.com and people would find it by typing "purses.com" in the address bar.

Now you definitely can NOT get something that people will just type in, and if you happen to be so lucky to find one, it will more than likely be found via google than someone hoping that your URL is what they'll need.

Wow, that was some rambling!

 

[3things]

When I needed something custom coded I went to elance (first time user) and firstly spent hours looking at potential workers who had good experience/knowledge in what I wanted done (scraping, API's, Wordpress plugins), read feedback, etc etc - so before I even posted the job, I knew exactly which worker I wanted to do it based on all that.

Also checked out what kind of price he worked for in previous jobs, and then just put up a flat rate priced job at similar money, detailed requirements etc. Put the job live, sent the guy an invite to it, and a week or so later had a perfectly coded WP plugin. (I'd spent about 3 weeks screwing around and failing to do what he did in a week - I'm capable of customising his work, but not capable of writing it all from scratch)

My own takeaways:
- Fully research workers who have the specific experience in what you're doing before you even post your job
- Invite your preferred worker(s) to your job
- Workers will do only what you ask them to do - if you want your code commented, specify it
- Never use hourly billing

Hope it worked out for OP in the end.